Method and device for rapid estimation of human core temperature with double time scale heat flow

Abstract

The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method and device, belong to temperature estimation technical field.The application discloses a kind of double time scale heat flow human core temperature rapid estimation method

Description

Technical Field

[0001] This invention relates to the field of temperature estimation technology, and in particular to a method and apparatus for rapid estimation of human core temperature based on dual timescale heat flow. Background Technology

[0002] Core body temperature is a key indicator for assessing the body's thermal balance, metabolic level, and physiological safety, and is widely used in various scenarios such as fever screening, activity monitoring, perioperative care, and intensive care. Compared to surface temperature, core temperature can accurately reflect the true thermal state of deep tissues and internal organs, and has higher clinical monitoring and health monitoring application value. Currently, core temperature measurement methods in clinical and routine testing are divided into invasive and non-invasive types, and each type has significant limitations and is difficult to adapt to the needs of long-term wearable continuous monitoring.

[0003] Dual-channel heat flow method is currently the mainstream technology for non-invasive core temperature monitoring. It involves setting up two sets of channels with different thermal conductivity characteristics on the body surface, collecting temperature data from both sides to construct a differentiated heat flow relationship, avoiding interference from the unknown thermal resistance of human tissue, and thus achieving core temperature inversion. Existing technologies mainly rely on differences in channel materials, thickness, and thermal resistance parameters to form heat flow paths, and mostly build temperature measurement models based on the assumption of steady-state heat transfer to complete core temperature calculations.

[0004] Existing dual-channel heat flow temperature measurement schemes only support core temperature calculation under steady-state conditions, lack transient dynamic inversion capabilities, and have poor environmental immunity, making measurement results susceptible to fluctuations caused by external disturbances, thus making it difficult to achieve high-precision and high-reliability continuous dynamic monitoring. Summary of the Invention

[0005] This invention provides a method and apparatus for rapid estimation of human core temperature based on dual timescale heat flow, in order to solve the technical problems of existing technologies that are limited by steady-state heat transfer assumptions, lack the ability to dynamically invert core temperature transiently, are sensitive to environmental disturbances, and are prone to fluctuations in measured values ​​during continuous monitoring, making it difficult to meet the requirements of high-precision and high-reliability dynamic monitoring.

[0006] In a first aspect, embodiments of the present invention provide a method for rapid estimation of human core temperature based on dual-timescale heat flux, comprising: Fast and slow heat conduction modules are attached to human skin to construct fast and slow channels respectively. A heat insulation control layer is set on the side of the fast and slow heat conduction modules away from the skin. The effective thermal conductivity of the fast and slow channels is calculated based on the thickness and cross-sectional area of ​​the fast and slow heat conduction modules respectively. Based on the thermal insulation control layer, the basic temperature information of the fast channel and the slow channel is collected. The natural transient identification stage is determined according to the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, the dual-channel dynamic consistency residual and environmental residual are constructed according to the basic temperature information, and the fast channel time constant, slow channel time constant and environmental heat transfer coefficient are estimated. Based on the fast channel time constant and the slow channel time constant, the fast channel and the slow channel are respectively subjected to dual time scale transient correction to obtain the equivalent corrected temperature difference of each channel, and the equivalent heat flux of each channel is reconstructed by combining the effective thermal conductance of the fast channel and the slow channel. Using the environmental heat transfer coefficient and base temperature information, the environmental reference heat flux corresponding to the fast channel and the slow channel is calculated respectively. The difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel is used to obtain the environmental boundary disturbance amount corresponding to each channel. The environmental boundary disturbance amount is compensated by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel. The transient core temperature inversion relationship is established, and the transient core temperature of the human body is calculated.

[0007] Secondly, embodiments of the present invention also provide a device for rapid estimation of human core temperature based on dual-timescale heat flux, comprising: The module is used to attach the fast heat conduction module and the slow heat conduction module to human skin, constructing fast channels and slow channels respectively, and setting a heat insulation control layer on the side of the fast heat conduction module and the slow heat conduction module away from the skin. The effective thermal conductivity corresponding to the fast channel and the slow channel is calculated based on the thickness and cross-sectional area of ​​the fast heat conduction module and the slow heat conduction module respectively. The estimation module is used to collect the basic temperature information of the fast channel and the slow channel based on the thermal insulation control layer, determine the natural transient identification stage according to the synchronous change characteristics of the basic temperature information, and construct the dual-channel dynamic consistency residual and environmental residual according to the basic temperature information and estimate the fast channel time constant, the slow channel time constant and the environmental heat transfer coefficient. The correction module is used to perform dual-time-scale transient correction on the fast channel and the slow channel according to the fast channel time constant and the slow channel time constant respectively to obtain the equivalent corrected temperature difference of each channel, and reconstruct the equivalent heat flux of each channel by combining the effective thermal conductance of the fast channel and the slow channel. The compensation calculation module is used to calculate the environmental reference heat flux corresponding to the fast channel and the slow channel respectively using the environmental heat transfer coefficient and the basic temperature information. The difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel is used to obtain the environmental boundary disturbance amount corresponding to each channel. The environmental boundary disturbance amount is compensated by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel and to establish the transient core temperature inversion relationship to calculate the transient core temperature of the human body.

[0008] Thirdly, embodiments of the present invention also provide an apparatus, comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the fast human core temperature estimation method for dual timescale heat flux as provided in the above embodiments.

[0009] Fourthly, embodiments of the present invention also provide a medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the rapid estimation method for human core temperature based on dual-timescale heat flux as provided in the above embodiments.

[0010] The present invention provides a method and apparatus for rapid estimation of human core temperature based on dual-timescale heat flux. A fast heat conduction module and a slow heat conduction module are attached to human skin to construct fast and slow channels, respectively. A heat insulation control layer is placed on the side of the fast and slow heat conduction modules away from the skin. The effective heat conductance corresponding to the fast and slow channels is calculated based on the thickness and cross-sectional area of ​​the fast and slow heat conduction modules, respectively. Based on the heat insulation control layer, the basic temperature information of the fast and slow channels is collected. A natural transient identification stage is determined based on the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, dual-channel dynamic consistency residuals and environmental residuals are constructed based on the basic temperature information, and the time constants of the fast and slow channels are estimated. The system employs a dual-timescale transient correction method based on the environmental heat transfer coefficient and the fast and slow channel time constants to obtain the equivalent corrected temperature difference for each channel. This is combined with the effective thermal conductance of the fast and slow channels to reconstruct the equivalent heat flux for each channel. Using the environmental heat transfer coefficient and baseline temperature information, the system calculates the environmental reference heat flux for both the fast and slow channels. The difference between the equivalent heat flux and the environmental reference heat flux for each channel is used to obtain the environmental boundary disturbance for each channel. A preset compensation coefficient is used to compensate for this disturbance, resulting in the compensated equivalent heat flux for each channel. A transient core temperature inversion formula is then established to calculate the transient core temperature of the human body. By combining a fast and slow dual-channel heat conduction structure with an external thermal insulation control layer, identifying dynamic heat transfer parameters using dual-channel residuals during the skin-contact transient phase, and reconstructing the equivalent heat flux using dual-timescale correction, along with compensation for environmental boundary disturbances, the system achieves direct inversion of the transient core temperature under dynamic human body conditions. This allows for high-precision body temperature estimation without waiting for a steady-state heat transfer state. Attached Figure Description

[0011] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0012] Figure 1 This is a flowchart of the rapid estimation method for human core temperature based on dual timescale heat flow provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the dual-timescale structure and the spatial arrangement of the thermal insulation control layer of the rapid estimation method for human core temperature based on dual-timescale heat flow provided in Embodiment 1 of the present invention. Figure 3 This is a schematic diagram of the method for rapid estimation of human core temperature based on dual timescale heat flow provided in Embodiment 1 of the present invention, which uses a robust sequential Monte Carlo algorithm for parameter estimation. Figure 4 This is a flowchart of the rapid estimation method for human core temperature based on dual timescale heat flow provided in Embodiment 2 of the present invention; Figure 5 This is a schematic diagram of the structure of the rapid human core temperature estimation device with dual time-scale heat flow provided in Embodiment 3 of the present invention; Figure 6 This is a schematic diagram of the device provided in Embodiment 4 of the present invention. Detailed Implementation

[0013] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0014] Example 1 Figure 1 This is a flowchart of a rapid estimation method for human core temperature based on dual-timescale heat flux provided in Embodiment 1 of the present invention. This embodiment is applicable to application scenarios where wearable non-invasive devices perform non-steady-state rapid and high-precision monitoring of human core temperature, and specifically includes the following steps: Step 110: Attach the fast heat conduction module and the slow heat conduction module to the human skin to construct the fast channel and the slow channel respectively. Set a heat insulation control layer on the side of the fast heat conduction module and the slow heat conduction module away from the skin. Calculate the effective thermal conductivity of the fast channel and the slow channel according to the thickness and cross-sectional area of ​​the fast heat conduction module and the slow heat conduction module respectively.

[0015] Relying solely on a single thermally conductive patch cannot separate the actual heat flow from environmental interference, nor can it construct a dual-channel inversion equation that eliminates the thermal resistance of unknown subcutaneous tissue. If the heat transfer response characteristics of the dual-channel modules are identical, differentiated transient thermal responses cannot be generated for parameter identification and heat flow decomposition. Furthermore, the lack of an external thermal insulation control layer allows ambient room temperature and air convection disturbances to directly intrude into the temperature measurement end, causing distortion in heat flow acquisition. Therefore, a hardware structure with differentiated fast and slow thermal conduction, combined with an external thermal insulation control layer, and quantitative calculation of effective thermal conductivity based on module geometry and material parameters, is the prerequisite physical condition for the entire algorithm model to be implemented and solved.

[0016] For example, such as Figure 2 The diagram shows a dual-timescale structure and the spatial arrangement of the thermal insulation control layer. A first heat conduction module (fast heat conduction module) and a second heat conduction module (slow heat conduction module) are arranged together on the same flexible substrate. They are adhered to the adjacent skin surface area via an adhesive layer, forming a stable thermal contact. The two modules respectively constitute the fast and slow channels. The heat conduction modules are preferably based on PDMS substrate. Differentiation can be achieved through at least two of the following: thermal conductivity, thickness, effective heat transfer cross-sectional area, volumetric heat capacity, and composite sandwich structure. This ensures that the equivalent thermal resistance and equivalent heat capacity of the two channels are unequal, achieving differentiation between fast and slow responses. On the side of the two heat conduction modules away from the skin, a continuous, full-coverage environmental thermal coupling control layer, i.e., a thermal insulation control layer, is made of a low thermal conductivity material with a much lower thermal conductivity than the substrate of the heat conduction modules. The thickness of the thermal insulation control layer is set to [thickness value missing]. and thermal conductivity is It can be done through the formula Calculate its own thermal resistance per unit area, and then add the convective heat transfer resistance between the outer surface of the module and the air. ( The combined thermal resistance on the ambient side is obtained by considering the convective heat transfer coefficient between the outer surface and the ambient air. This allows for the calculation of the equivalent environmental heat transfer coefficient. In the complete process of calculating the effective thermal conductivity of the dual-channel system, the thermal conductivity of the materials used in the fast and slow thermal conduction modules is first extracted separately. Module body thickness and the effective heat transfer cross-sectional area perpendicular to the direction of heat conduction Then follow the formula The one-dimensional axial equivalent thermal resistance of the fast and slow heat conduction modules is calculated separately, where i is the thermal channel number, with a value of 1 corresponding to the fast heat conduction module and a value of 2 corresponding to the slow heat conduction module. Then, the equivalent thermal resistances of the two modules are directly calculated by taking the reciprocal, and finally the effective thermal conductances of the fast and slow channels are obtained respectively.

[0017] This structure, on the one hand, relies on differentiated fast and slow dual channels to generate different transient thermal response characteristics, creating the core hardware conditions for eliminating the equivalent thermal resistance of unknown subcutaneous tissue in the human body through simultaneous equations. The outer full-coverage thermal insulation control layer effectively extends the path of environmental temperature fluctuations to the internal temperature measurement nodes, significantly reducing the direct impact of external temperature fluctuations and air convection on the temperature measurement points, and greatly improving the stability and purity of the original temperature acquisition data. On the other hand, the effective thermal conductivity of the dual channels, obtained through quantitative calculation of material properties and geometric dimensions, provides a rigorous physical constraint boundary for the first-order lumped parameter thermal model, ensuring the physical rationality of the subsequent channel time constant identification and equivalent heat flow reconstruction calculation results.

[0018] Step 120: Based on the thermal insulation control layer, collect the basic temperature information of the fast channel and the slow channel. Determine the natural transient identification stage according to the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, construct the dual-channel dynamic consistency residual and environmental residual according to the basic temperature information and estimate the fast channel time constant, the slow channel time constant and the environmental heat transfer coefficient.

[0019] By relying on the thermal insulation control layer to shield environmental disturbances, complete basic temperature data from multiple points are collected. The transient identification start window is automatically determined based on the synchronous change characteristics of skin temperature. This avoids the errors caused by manual marking and ensures that heat transfer parameters are identified only using the real transient thermal excitation generated by skin contact. This avoids environmental temperature fluctuations from triggering parameter solving and ensures the accuracy and reliability of subsequent model parameter identification.

[0020] For example, multiple temperature sensors are used to collect temperatures. An outer thermal insulation layer is installed on the side of the fast and slow heat conduction modules furthest from the skin. This thermal insulation layer acts as a thermal buffer only for the temperature measurement points on the fast and slow environmental sides, reducing interference from external environmental fluctuations. Based on this structure, five types of basic temperature information are stably collected: fast-channel skin-contact temperature, fast-channel environmental temperature, slow-channel skin-contact temperature, slow-channel environmental temperature, and external environmental temperature. During data preprocessing, at least one of the following methods can be used to denoise and filter all collected temperature time-series signals: moving average filtering, median filtering, low-pass filtering, and local polynomial filtering. Then, the rate of change of the fast-channel and slow-channel skin-contact temperatures compared to the previous sampling time is calculated using adjacent sampling differences, central differences, or local fitting differentiation. Simultaneously, the axial temperature difference between the skin-contact end and the environmental end of each of the two thermal channels and its rate of change can be calculated as auxiliary judgment criteria. In terms of judgment logic, when it is detected that the temperature rise and fall directions of the fast channel skin-contact side temperature and the slow channel skin-contact side temperature are completely consistent, and the rate of change of the skin-contact side temperature of each of the two channels exceeds the preset attachment response threshold, and this condition can be maintained continuously for a preset complete sampling period, it can be determined that the fast and slow heat conduction modules have achieved reliable contact with human skin. The current sampling time that meets the condition is defined as the natural attachment time. The subsequent natural transient identification stage is defined by taking this natural attachment time as the time starting point. After the device switches from being suspended and not attached to being stably attached to the skin, the thermal conduction boundary conditions between the skin and the dual-channel heat conduction modules change abruptly, which will naturally form an effective transient thermal excitation that can be used by the algorithm to solve for the fast channel time constant, the slow channel time constant and the environmental heat transfer coefficient.

[0021] After defining a pure and effective natural transient identification stage, a comprehensive objective function is constructed by combining dual-channel dynamic consistency residuals and environmental residuals. The parameter identification is then completed using a robust sequential Monte Carlo algorithm. This approach can eliminate the unknown human body heat flow that cannot be directly measured and avoid the interference of environmental disturbances on parameter solving. Furthermore, it can robustly solve the dual-channel time constant and environmental heat transfer coefficient under physical constraints, providing accurate model kernel parameters for subsequent dual-timescale heat flow correction, environmental disturbance compensation, and transient core temperature inversion.

[0022] For example, such as Figure 3To illustrate the parameter estimation using the robust sequential Monte Carlo algorithm, within the natural transient identification phase defined starting from the natural attachment moment, the axial temperature difference of the fast channel and the slow channel are calculated based on the five-channel basic temperature information that has been synchronously acquired and filtered for noise reduction. Then, the rate of change corresponding to the two sets of axial temperature differences is solved. Before generating candidate particles, based on known heat transfer relationships and the material thermal properties, structural design dimensions, manufacturing tolerances, design allowable attachment pressure range, insulation control layer parameters, and environmental conditions of the fast and slow heat conduction modules, the physical feasible region of the fast channel time constant, slow channel time constant, dual-channel gain ratio coefficient, and equivalent heat transfer coefficient on the environmental side is pre-determined. For the first... Each heat conduction channel equates the main transient heat transfer processes of the heat conduction module, encapsulation layer, adhesive layer, and skin-contact interface to a first-order thermal resistance and thermal capacity network. For the first... A solid heat transfer layer with axial thermal resistance According to Fourier one-dimensional thermal conductivity Calculation, where The thickness of this layer along the main heat transfer direction. The thermal conductivity of this layer of material is... This is the effective heat transfer area. (Number) The contact thermal resistance of the channel skin interface is expressed as , For skin-contact heat transfer coefficient, To ensure effective skin contact area. The first [unit / entity] involved in transient heat storage. Heat capacity of each material layer ,in , and These are the material density, specific heat capacity, and effective volume, respectively. Based on the first-order thermal resistance, the energy conservation relationship of the heat capacity network, the thermophysical parameters of each layer of material, and the structural dimensions, the... The reference time constant of the channel satisfies ,in and The first The equivalent thermal resistance and equivalent heat capacity of the channel, where the equivalent thermal resistance relationship satisfies The equivalent heat capacity relationship satisfies Then, by taking values ​​for the material's thermal conductivity, density, specific heat capacity, structural thickness, effective area, and skin-contact heat transfer coefficient within a certain range, and maintaining the actual physical correlation between each parameter, the lower limit of the time constant can be calculated. and upper limit of time constant The first time constant is determined based on the maximum relative deviation of its upper and lower bounds from the reference time constant. Channel deviation coefficient ,Right now In the formula, This is used to characterize the maximum relative deviation of the time constant corresponding to changes in material thermophysical properties, structural manufacturing tolerances, and design contact conditions, thereby determining the first... The physical feasible region of the channel time constant is , ,in and The first The physical lower and upper limits of the channel time constants are defined, and a dual-channel fast / slow constraint is enforced, stipulating that the fast channel time constant is less than the slow channel time constant. This prevents the response characteristics of the two channels from being too similar, resulting in insufficient dynamic residual identification and ineffective parameter differentiation. For the dual-channel candidate gain scaling factor, in the... At the temperature sampling time, the first The candidate dynamic correction temperature difference for each channel is In the formula, For the first The first channel in the Candidate dynamic correction temperature difference at each sampling time. For the first The first channel in the Axial temperature difference at each sampling time, For the first The first channel in the Each sampling time corresponds to a candidate time constant for a candidate particle. For the first The first channel in the The rate of change of axial temperature difference at each sampling time. Based on the candidate equivalent heat flux relationship. ,in For the first The first channel in the Candidate equivalent heat flux at each sampling time, For the first The effective thermal conductivity of each channel. Since the fast and slow channels are arranged in adjacent body surface areas, under the same or nearly the same local human body heat input conditions, the candidate equivalent heat flux obtained by the two channels after correction for their respective transient characteristics and effective thermal conductivity should be approximately the same. Therefore, Thus, the theoretical gain ratio of the two channels is obtained. ,in This is the theoretical gain ratio calculated based on structural parameters such as material thermal conductivity, thickness, and area. For multilayer series structures, there is... Therefore, based on the allowable range of material thermal conductivity and structural dimensions, it can be directly obtained that... and Further determination and Thus formed The physically feasible region. For the equivalent heat transfer coefficient on the environmental side... The thermal resistance per unit area of ​​the thermal insulation control layer is combined with the convective thermal resistance per unit area from its outer surface to the ambient air in a series relationship, wherein the thermal resistance per unit area of ​​the thermal insulation control layer is... The convective thermal resistance per unit area on the air side is , and These are the thickness and thermal conductivity of the insulation control layer, respectively. The convective heat transfer coefficient between the outer surface of the insulation layer and the ambient air is given; therefore, the equivalent heat transfer coefficient on the ambient side satisfies... Obtained from the range of environmental parameters and Then, by combining the thickness of the insulation control layer and the thermal conductivity, it can be determined and By integrating all constraints, a complete feasible domain set is obtained. Within this set, all initial candidate particle sets are generated, and particle state prediction is performed. At the k-th temperature sampling time, based on the initial candidate particles from the previous sampling time, a small random perturbation is introduced by superimposing zero-mean small-amplitude process noise to generate candidate particles for the current time. Considering that the dual-channel heat conduction structure, the thermal resistance of the attachment contact, and the external convection environment remain stable during a single complete human wearing process, the fast-channel candidate time constant, the slow-channel candidate time constant, the dual-channel candidate gain ratio coefficient, and the candidate environment heat transfer coefficient are all quasi-static slow time-varying parameters. Based on this, the perturbation application logic can be optimized by uniformly adding this noise only after each round of particle resampling, ensuring the diversity of particle population traversal while avoiding unnecessary parameter fluctuations. After the predicted particles are generated, hard constraint verification is carried out immediately. On the one hand, it is verified that the values ​​of the four parameters are all positive. On the other hand, the inherent heat transfer logic that the fast channel time constant must be less than the slow channel time constant is verified. For non-compliant particles that do not meet the constraint conditions, they are directly truncated and corrected to the boundary of the feasible region, or compliant particles are regenerated within the physical feasible region to eliminate invalid parameter combinations that violate the heat transfer mechanism from the source.

[0023] For each qualified predicted particle, the dual-channel dynamic response residual is calculated, and then substituted into the formula to solve for the dynamic consistency residual corresponding to a single particle. ,in, Let be the candidate time constant for the fast channel at sampling time k. Let be the rate of change of axial temperature difference in the fast channel at sampling time k. Let be the axial temperature difference of the fast channel at sampling time k. Let be the candidate gain scaling factor for the dual channels at sampling time k. Let be the candidate time constant for the slow channel at sampling time k. Let be the rate of change of axial temperature difference in the slow channel at sampling time k. Let k be the axial temperature difference of the slow channel at sampling time k. This represents the dual-channel dynamic consistency residual at sampling time k. The underlying calculation logic utilizes two sets of first-order heat transfer dynamic models. ,in The common thermal excitation inputted by the human body to the dual channels, For fast-channel equivalent thermal response gain, For the slow-channel equivalent thermal response gain, based on the candidate equivalent heat flux relationship The heat flux experienced by both channels is approximately equal to the common input heat flux. Therefore, through , and Comparison yields Similarly... For multi-layer structures, the total equivalent thermal resistance, including the thermally conductive module layer, encapsulation layer, and contact thermal resistance, is calculated, and then the total equivalent thermal resistance is used to calculate... ,thus and It is approximately equal to the reciprocal of the total equivalent thermal conductivity of the corresponding heat-conducting module layer. Definition The gain ratio of the fast and slow channels in the actual dynamic model is the real operating value identified through actual temperature data. This is achieved through parameters... The dual-channel thermal response amplitude is uniformly converted to the same dimension to eliminate the influence of unknown human body input heat flow. When the candidate parameters are infinitely close to the real physical parameters, the dual-channel equivalent thermal response is highly consistent and the dynamic consistency residual approaches zero. The larger the parameter deviation, the higher the residual value. This is used to quantify the degree of matching between each set of parameters and the dual-channel transient heat transfer law.

[0024] Meanwhile, the environmental boundary residuals are solved for the same batch of particles, and the dual-channel candidate equivalent heat flux is reconstructed by relying on the two sets of time constants carried within the particles. ,in, For the first Candidate equivalent heat flux at sampling time k for each channel For the first Effective thermal conductivity of each channel For the first The axial temperature difference at each of the k sampling times in each channel. For the first Candidate time constants for k sampling times of each channel For the first The rate of change of axial temperature difference at sampling time k in each channel. Then, relying on the environmental side heat transfer cross-sectional area of ​​the two thermal channels. and The ambient temperature was calculated. ,in, Let k represent the ambient temperature at sampling time k. The cross-sectional area of ​​the fast heat conduction module, This refers to the cross-sectional area of ​​the slow-conducting heat module. Let k be the ambient temperature of the fast channel at sampling time k. Let be the ambient temperature of the slow channel at sampling time k. Simultaneously, candidate equivalent heat fluxes for both channels are weighted, and the ambient residual is constructed by combining the candidate ambient heat transfer coefficients with the ambient temperature. ,in, Let k be the environmental residual at sampling time k. Let be the heat transfer coefficient of the candidate environment at sampling time k. The cross-sectional area of ​​the fast heat conduction module, This refers to the cross-sectional area of ​​the slow-conducting heat module. Let k represent the ambient temperature at sampling time k. Let k be the ambient temperature at sampling time k. Let k be the weighted values ​​of the candidate equivalent heat fluxes for the fast and slow channels at sampling time k, satisfying the following relationship: , For the first Candidate equivalent heat flux at sampling time k for each channel For the first Candidate equivalent heat fluxes at sampling time k in each channel. A smaller environmental residual indicates a higher degree of fit between the equivalent heat transfer coefficient corresponding to the particle and the actual external heat dissipation conditions of the equipment, thus completing the dual constraint determination of internal dynamic characteristics and external boundary conditions of heat transfer. Normalization is performed on the two types of residuals, and a comprehensive residual cost function is constructed. To eliminate the weight imbalance problem caused by the inconsistency in dimensions and numerical fluctuation range between dynamic and environmental residuals, the normal fluctuation scale of the residuals obtained through offline calibration is used. and Scaling and normalizing are performed on the two types of residuals respectively, and then weighted by the coefficients. and Weighted fusion yields the comprehensive residual Among them, the weighting coefficient and The calibration was pre-obtained through offline calibration. The calibration process involved constructing a calibration device using the same fast thermal conductivity module, slow thermal conductivity module, thermal insulation control layer, and encapsulation structure as the actual measuring device. Calibration tests were conducted at different ambient and core temperatures, simultaneously collecting data on the fast channel skin-side temperature, fast channel ambient temperature, slow channel skin-side temperature, slow channel ambient temperature, ambient temperature, and reference core temperature. Based on the calibration dataset under normal operating conditions, the dual-channel dynamic consistency residual and environmental residual were calculated, and normalization was performed according to the normal fluctuation range of both types of residuals. Subsequently, within a preset constraint range (0 < ... <1, 0< <1, + =1) Multiple candidate combinations of dynamic consistency residual weights and environmental residual weights are set within the framework. Each candidate weight combination is substituted into the robust sequential Monte Carlo algorithm to sequentially identify and estimate the fast-channel time constant, slow-channel time constant, dual-channel gain scaling factor, and environmental heat transfer coefficient, and further solve for the human core temperature. Finally, the calculated core temperature results are compared with the reference core temperature, and the candidate combination with the smaller core temperature estimation error and lower output fluctuation is selected as the dynamic consistency residual weight. Environmental residual weights The final calibration result. In the formula, The combined residual at sampling time k is... Let K be the dynamic consistency residual of the two channels at sampling time k. The environmental residual at sampling time k is the overall residual. The lower the overall value of the residual, the better the overall fitting effect is, as it means that the particle parameters of this set simultaneously satisfy the heat transfer consistency of the dual-channel internal environment and the heat dissipation boundary conditions of the external environment.

[0025] The particle weights are iteratively updated based on the comprehensive residual. A heavy-tailed likelihood calculation method with outlier resistance is introduced to obtain the fitting degree of each particle. Two adjustable robust parameters are used to adjust the fault tolerance. The smaller the comprehensive residual, the higher the likelihood matching degree of the particle. The particle weights retained from the previous iteration are multiplied by the likelihood matching degree calculated in the current iteration to obtain the temporary weights without normalization. The temporary weights of all particles are then summed, and global normalization is performed by dividing the temporary weight of each particle by the total weights to obtain the official effective weights for the current iteration. The heavy-tailed likelihood calculation rule can effectively avoid the problem of high-quality particle weights being directly reduced to zero due to abnormal sampling points such as instantaneous loosening of the wearer, single-point temperature measurement noise, and sudden changes in environmental airflow, which greatly improves the robustness of the identification algorithm. Subsequently, particle degradation detection and resampling operations are carried out. The current particle weight is calculated by taking the reciprocal of the sum of the squares of the normalized weights of all particles. The number of effective particles in the set is determined. When the number of effective particles is lower than the system's preset threshold, it is determined that the particles have suffered severe weight degradation and the vast majority of particles have lost their effective contribution. Then, based on the current weight distribution of the particles, system resampling, hierarchical resampling, or residual-guided resampling is performed. High-weight, high-quality fitting particles are copied and retained, while invalid particles with weights close to zero are completely eliminated. After resampling, the weights of all particles are uniformly reset to equal values, and a small random perturbation is added again. At the same time, the physical feasible region constraint is checked to avoid all particles collapsing to the same value in the later stages of iteration and losing the ability to optimize parameters. Then, the posterior estimate of the parameters at the sampling time of this round is solved. The normalized particle weights are used as weighting coefficients, and the four parameters of all compliant particles are weighted and averaged. The four results after weighted averaging correspond to the average fast channel time constant, slow channel time constant, dual-channel gain ratio, and equivalent environmental heat transfer coefficient after convergence, respectively.

[0026] Step 130: Perform dual-time-scale transient correction on the fast channel and slow channel according to the fast channel time constant and the slow channel time constant respectively to obtain the equivalent corrected temperature difference corresponding to each channel, and reconstruct the equivalent heat flux corresponding to each channel by combining the effective thermal conductance corresponding to the fast channel and the slow channel.

[0027] The dual-channel heat transfer structure has first-order inertial heat capacity characteristics. The original axial temperature difference obtained by temperature measurement has obvious dynamic lag and cannot be directly equated to the actual heat transfer driven temperature difference under unsteady state. Direct use will cause the heat flow calculation to deviate from the actual working condition, ultimately resulting in a significant decrease in the accuracy of the human body core temperature inversion. Therefore, it is necessary to use the identified fast and slow channel time constants to implement dual time scale transient correction to compensate for the lag bias, and then rely on the effective thermal conduction of the channels to complete the equivalent heat flow reconstruction, so as to provide accurate unsteady state heat input for subsequent body temperature inversion.

[0028] For example, based on the fast channel time constant and slow channel time constant output after algorithm iteration convergence, dual-time-scale transient correction calculations are performed on the fast channel and slow channel respectively, thereby solving for the equivalent corrected temperature difference corresponding to the k-th sampling time for each of the two channels. The single-channel dual-time-scale transient correction calculation formula is as follows: ,in, For the first The equivalent corrected temperature difference at each of the k sampling times of each channel. For the first The axial temperature difference at each of the k sampling times in each channel. To estimate the obtained first The time constant of each channel at k sampling times For the first The axial temperature difference change rate at sampling time k for each channel is used to introduce a time constant to compensate for the temperature difference change rate, thus offsetting the dynamic response lag caused by the heat capacity of the heat transfer system and restoring the true heat transfer-driven temperature difference under unsteady-state conditions. After calculating the equivalent corrected temperature difference for the fast and slow channels respectively, the equivalent heat flux of each channel is reconstructed by combining the effective thermal conductivity parameters of the two channels. The formula for calculating the equivalent heat flux of a single channel is as follows: ,in, For the first Equivalent heat flux at k sampling times for each channel For the first Effective thermal conductivity of each channel For the first The equivalent corrected temperature difference at each of the k sampling times of each channel, the equivalent heat flux can accurately characterize the unsteady stage of the transient thermal excitation of the attachment. The actual thermal response of the dual channels after dynamic hysteresis correction can be used as a reliable thermal boundary input for the back-end human core temperature inversion calculation, reducing the calculation error caused by transient heat transfer hysteresis from the source.

[0029] Step 140: Calculate the environmental reference heat flux corresponding to the fast channel and the slow channel using the environmental heat transfer coefficient and the base temperature information respectively. Subtract the equivalent heat flux corresponding to each channel from the environmental reference heat flux corresponding to each channel to obtain the environmental boundary disturbance amount corresponding to each channel. Compensate the environmental boundary disturbance amount by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel and establish the transient core temperature inversion relationship to calculate the transient core temperature of the human body.

[0030] The dual-channel system experiences convective heat loss from the outside air. The equivalent heat flow obtained solely based on internal temperature difference correction does not account for boundary disturbances caused by environmental heat dissipation. Directly using this for core temperature inversion will overestimate the actual effective heat conducted inward from the body, resulting in a lower-than-expected body temperature and poorer dynamic tracking. Therefore, it is necessary to first calculate the environmental reference heat flow emitted outward from the fast and slow channels separately based on the identified environmental heat transfer coefficients, extract the environmental boundary disturbances, and correct the original equivalent heat flow using a fixed compensation coefficient. After removing external heat dissipation interference, the true effective heat flow input into the body is obtained, ensuring the accuracy and physical plausibility of the transient core temperature inversion calculation.

[0031] For example, based on the environmental equivalent heat transfer coefficient estimated by particle filtering posterior estimation, and combined with the real-time acquired environmental temperature and the fast-channel environmental side temperature, the quantitative calculation of the reference heat flux on the fast-channel and slow-channel environmental sides is completed respectively. The formula for calculating the reference heat flux on the fast-channel environmental side at the k-th sampling time is as follows: ,in, The ambient reference heat flux of the hot channel at sampling time k. The environmental heat transfer coefficient is estimated at sampling time k. The cross-sectional area of ​​the fast heat conduction module, Let k be the ambient temperature of the fast channel at sampling time k. Let K be the ambient temperature at sampling time k. The corresponding slow-channel ambient-side reference heat flux calculation formula is: ,in, The environmental reference heat flux for the slow channel at sampling time k. The environmental heat transfer coefficient is estimated at sampling time k. This refers to the cross-sectional area of ​​the slow-conducting heat module. Let k be the ambient temperature of the slow channel at sampling time k. Let k be the ambient temperature at sampling time. Both channels quantify the heat loss to the outside using Newton's convection heat dissipation formula. After obtaining the ambient reference heat flux for each channel, the equivalent heat flux of the channel, reconstructed through dual-timescale hysteresis correction, is subtracted from the corresponding channel's ambient reference heat flux. The difference is the ambient boundary perturbation for a single channel, representing the deviation in heat loss caused by external convection heat dissipation to the heat transfer link. Subsequently, a pre-calibrated ambient boundary compensation coefficient for each channel is introduced to perform closed-loop correction on the perturbation. The ambient boundary compensation coefficient... Pre-calibration was performed offline. The calibration process involved constructing a calibration device using the same fast thermal conductivity module, slow thermal conductivity module, thermal insulation control layer, and encapsulation structure as the actual measuring device. Calibration tests were conducted at different ambient and core temperatures, simultaneously collecting data on the skin-side temperature of the fast channel, the ambient temperature of the fast channel, the skin-side temperature of the slow channel, the ambient temperature of the slow channel, the ambient temperature, and the reference core temperature. Based on the collected temperature data, the equivalent heat flux of the fast and slow channels without environmental compensation, the reference heat flux on the ambient side, and the corresponding environmental boundary perturbations were calculated. This was performed within a preset physically feasible range (0 < 0). <1) Multiple candidate values ​​for environmental boundary compensation coefficients are set up. Each candidate value is used to compensate for the environmental boundary disturbances in the fast and slow channels, yielding the compensated equivalent heat flux for each candidate value. The compensated equivalent heat fluxes for the fast and slow channels are substituted into the core temperature inversion formula, and the inverted core temperature is compared with the reference core temperature. Candidate values ​​with smaller core temperature errors and lower output fluctuations are selected as the environmental boundary compensation coefficients for the corresponding channels. The environmental boundary compensation coefficients for the fast and slow channels are calibrated independently. (Using the formula...) ,in, The first sample after compensation at time k The equivalent heat flux of each channel, For the first Effective thermal conductivity of each channel For sampling time k, the first... The equivalent corrected temperature difference for each channel For the first Preset environmental boundary compensation coefficients for each channel. For sampling time k, the first... The calculation of environmental boundary disturbances for each channel completes the quantitative elimination of environmental heat dissipation disturbances.

[0032] Optionally, this embodiment also performs common-mode-differential-mode decomposition on the compensated fast-channel equivalent heat flux and slow-channel equivalent heat flux. First, a matching coefficient is calculated based on the effective thermal conductivity ratio of the two channels and the estimated candidate gain scaling factor of the two channels. The calculation formula is as follows: In the formula and The effective thermal conductivity of the two heat channels are respectively. The estimated dual-channel candidate gain scaling factor is used to uniformly compensate for the inherent dynamic gain and thermal conductivity differences between the two channels. The common-mode thermal response and differential-mode thermal response are solved based on the matching coefficient. The common-mode thermal response calculation formula is as follows: The formula for calculating differential mode thermal response is: ,in Let k be the equivalent heat flux of the fast channel after compensation at sampling time k. The equivalent heat flux of the slow channel after compensation at sampling time k is given. The common-mode thermal response characterizes the intensity and trend of the actual local heat input to the human body jointly picked up by the two thermal channels. When the amplitude or rate of change of the common-mode thermal response reaches a preset threshold, it is determined that the effective heat input to the human body has been sufficiently established, and the system can maintain continuous analysis and iterative updates of each parameter to be estimated. If the amplitude of the common-mode thermal response is too low, it is determined that neither channel has effective temperature measurement input, and it is not suitable to continue performing parameter posterior updates to avoid parameter drift caused by invalid iteration. The differential-mode thermal response characterizes the remaining inconsistency deviation after the characteristic matching of the two channels. Fault and interference source tracing can be completed based on the differential-mode change characteristics. A synchronous abrupt change in the heat flux of both channels and a significant synchronous fluctuation in the common-mode indicate a change in the sensor's skin contact state. A sudden change in the heat flux of only one channel or baseline drift indicates an abnormality in the hardware acquisition of that channel. A slow increase in the differential mode and a strong correlation with the local temperature gradient indicate lateral heat conduction and dissipation interference from the skin. A large jump in the differential mode accompanied by a large change in the ambient temperature indicates a rapid disturbance in the external ambient temperature. To quantify the severity of the anomaly, a normalized differential modulus is further constructed. In the formula For the normalized difference modulus, For differential mode thermal response, To obtain the preset differential mode threshold after calibration, the calibration method involves building a calibration device using the same fast and slow heat conduction modules, thermal insulation control layer, and encapsulation structure as the actual device. Data from two channels are collected under different ambient and core temperature conditions, and the differential mode thermal response is calculated for each normal sampling moment. Statistics on all normal operating conditions The distribution is taken from 95% of the normal data. The value that will not be exceeded is used as Finally, a batch of normal operating condition datasets that were not involved in the threshold calculation were retrieved to verify the calibration results, ensuring that the vast majority of normal operating condition data met the requirements. When the normalized differential modulus is less than or equal to 1, it is considered normal operation, and the reliability of dual-channel data is fully preserved. When the normalized differential modulus is greater than 1 but less than or equal to the first differential modulus threshold, it is judged as a slight anomaly. The weight of the abnormal channel in heat flow fusion and common mode calculation is reduced according to the preset attenuation coefficient to weaken the interference. The first differential modulus threshold is obtained through offline calibration, using the same fast and slow heat conduction modules, thermal insulation control layer, and encapsulation structure as the actual device to build the calibration device. Under different ambient temperatures and simulated core temperatures, mild abnormal disturbances such as weak airflow, slight ambient temperature jumps, and slight changes in attachment pressure, as well as moderate abnormal disturbances such as strong airflow, large ambient temperature jumps, and significant changes in attachment state, are applied, and dual-channel heat flow data and reference core temperature are collected simultaneously. For each abnormal sampling time, the corresponding differential modulus thermal response is calculated, and the calibrated differential modulus threshold is used. Normalization is performed to obtain the normalized differential modulus. Then, a weight reduction operation is performed. Based on whether the core temperature error calculated after reducing the weight of the abnormal channels is within the allowable range of ±0.3℃, the entire calibration dataset is divided into mildly abnormal data and moderately abnormal data. The distribution intervals of the normalized differential modulus corresponding to the mildly and moderately abnormal samples are statistically analyzed. A first differential modulus threshold is selected between the higher interval of the normalized differential modulus range for mildly abnormal samples and the lower interval of the normalized differential modulus range for moderately abnormal samples. When the data intervals of the two types of abnormal samples overlap, the value that minimizes the number of misclassified mildly and moderately abnormal samples is selected as the final first differential modulus threshold. After calibration, the mildly and moderately abnormal datasets that were not involved in the threshold calculation are retrieved to verify the effectiveness of the threshold. When the normalized differential modulus is greater than the first differential modulus threshold but less than or equal to the second differential modulus threshold, it is judged as a moderate anomaly. Real-time updates of core identification parameters such as the fast and slow channel time constants, gain ratio coefficients, and environmental heat transfer coefficients are paused, and the calculation continues using the most recently converged and reliable parameters. The second differential modulus threshold is also pre-calibrated offline, and the calibration device is built using the same fast and slow heat conduction modules, thermal insulation control layers, and encapsulation structures as the actual device. Under different ambient temperatures and simulated core temperatures, moderate anomalies such as strong airflow, large ambient temperature jumps, and significant changes in adhesion state, as well as severe anomalies such as strong direct airflow and severe sensor offset, are applied, and dual-channel heat flux data and reference core temperature are collected simultaneously. For each anomaly sampling moment, the differential modulus thermal response and normalized differential modulus are calculated, and the most recently converged and reliable identification parameters are retrieved to continue calculating the human core temperature at the current moment. The error between the calculated core temperature and the reference core temperature is compared, and the calculation is based on whether the error exceeds the maximum allowable error range of ±0.5℃ and whether the error falls within... The calibration dataset is divided into severely anomalous and moderately anomalous data. The distribution intervals of the normalized difference modulus for each type of anomalous sample are statistically analyzed. A second difference modulus threshold is determined between the interval with higher normalized difference modulus for moderate anomalous samples and the interval with lower normalized difference modulus for severe anomalous samples. When the intervals of the two types of anomalous data overlap, the value that minimizes the number of misclassified moderate and severe anomalous samples is selected as the final second difference modulus threshold. After calibration, validation is performed using moderate and severe anomalous datasets that were not involved in threshold calculation. When the normalized difference modulus is greater than the second difference modulus threshold, it is judged as a severe anomalous sample, the algorithm stops outputting valid core temperature measurement results, and a low-confidence status is marked externally.

[0033] For example, using the fast-channel equivalent heat flux and slow-channel equivalent heat flux after dynamic correction and environmental boundary disturbance compensation, and simultaneously introducing the fast-channel skin-to-skin temperature and slow-channel skin-to-skin temperature collected at the same sampling time, a transient core temperature inversion calculation formula is established to directly solve for the transient core temperature of the human body at the k-th sampling time. The specific calculation formula is as follows: ,in, Let k be the transient core temperature of the human body at sampling time k. Let k be the equivalent heat flux of the fast channel after compensation at sampling time k. Let k be the equivalent heat flux of the slow channel after compensation at sampling time k. For the fast-channel skin-contact temperature at sampling time k, The slow channel temperature at sampling time k is the temperature on the skin side. The calculation method fully utilizes the reliable heat flow data after hysteresis compensation, environmental heat dissipation elimination, and channel matching correction mentioned above. It avoids the calculation deviation caused by temperature measurement inertial hysteresis, external heat dissipation loss, and inconsistency of dual-channel characteristics. It can output high-precision transient core temperature results in real time according to the dynamic changes of human body heat production, which serves as the final output of the entire detection method.

[0034] In this embodiment, fast and slow heat conduction modules are attached to human skin to construct fast and slow channels respectively. A heat insulation control layer is placed on the side of the fast and slow heat conduction modules away from the skin. The effective thermal conductivity of the fast and slow channels is calculated based on the thickness and cross-sectional area of ​​the fast and slow heat conduction modules. Based on the heat insulation control layer, the basic temperature information of the fast and slow channels is collected. A natural transient identification stage is determined based on the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, dual-channel dynamic consistency residuals and environmental residuals are constructed based on the basic temperature information, and the time constants of the fast and slow channels and the environmental heat transfer coefficient are estimated. Based on the fast channel... The time constant and slow channel time constant are used to perform dual-timescale transient corrections on the fast and slow channels to obtain the equivalent corrected temperature difference for each channel. The equivalent heat flux for each channel is then reconstructed by combining the effective thermal conductance of the fast and slow channels. Using the environmental heat transfer coefficient and baseline temperature information, the environmental reference heat flux for the fast and slow channels is calculated. The difference between the equivalent heat flux and the environmental reference heat flux for each channel is used to obtain the environmental boundary disturbance for each channel. A preset compensation coefficient is used to compensate for the environmental boundary disturbance, resulting in the compensated equivalent heat flux for each channel. A transient core temperature inversion formula is then established to calculate the transient core temperature of the human body. By combining a fast and slow dual-channel heat conduction structure with an external thermal insulation control layer, identifying dynamic heat transfer parameters using dual-channel residuals during the skin-contact transient phase, and reconstructing the equivalent heat flux with dual-timescale correction and compensation for environmental boundary disturbances, the transient core temperature under dynamic human body conditions can be directly inverted and solved, achieving high-precision body temperature estimation without waiting for a steady-state heat transfer state.

[0035] Example 2 Figure 4 This is a flowchart of a rapid estimation method for human body core temperature based on dual timescale heat flux provided in Embodiment 2 of the present invention. This embodiment is an optimization based on the above embodiment. After the step of compensating the environmental boundary disturbance by using a preset compensation coefficient to obtain the equivalent heat flux corresponding to each channel after compensation and establishing the transient core temperature inversion relationship, and calculating the human body transient core temperature, it further includes: performing a difference calculation on the compensated fast channel equivalent heat flux and slow channel equivalent heat flux; when the difference is less than a preset threshold, marking the human body transient core temperature as a low confidence state, and selecting the most recently sampled reliable human body transient core temperature as the effective human body transient core temperature at the current moment.

[0036] See Figure 4 A rapid method for estimating human core temperature based on dual-timescale heat flux includes: Step 210: Attach the fast heat conduction module and the slow heat conduction module to the human skin to construct fast channels and slow channels respectively. Set a heat insulation control layer on the side of the fast heat conduction module and the slow heat conduction module away from the skin. Calculate the effective thermal conductivity of the fast channel and the slow channel according to the thickness and cross-sectional area of ​​the fast heat conduction module and the slow heat conduction module respectively.

[0037] Step 220: Based on the thermal insulation control layer, collect the basic temperature information of the fast channel and the slow channel. Determine the natural transient identification stage based on the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, construct the dual-channel dynamic consistency residual and environmental residual based on the basic temperature information and estimate the fast channel time constant, the slow channel time constant and the environmental heat transfer coefficient.

[0038] Step 230: Perform dual-time-scale transient correction on the fast channel and slow channel according to the fast channel time constant and the slow channel time constant respectively to obtain the equivalent corrected temperature difference corresponding to each channel, and reconstruct the equivalent heat flux corresponding to each channel by combining the effective thermal conduction corresponding to the fast channel and the slow channel.

[0039] Step 240: Calculate the environmental reference heat flux corresponding to the fast channel and the slow channel using the environmental heat transfer coefficient and the base temperature information respectively. Divide the equivalent heat flux corresponding to each channel with the environmental reference heat flux corresponding to each channel to obtain the environmental boundary disturbance amount corresponding to each channel. Compensate the environmental boundary disturbance amount by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel and establish the transient core temperature inversion relationship to calculate the transient core temperature of the human body.

[0040] Step 250: Perform a difference calculation on the compensated fast-channel equivalent heat flux and slow-channel equivalent heat flux. When the difference is less than a preset threshold, mark the human transient core temperature as a low confidence state, and select the most recently sampled confidence human transient core temperature as the current effective human transient core temperature.

[0041] The denominator of the dual-channel core temperature inversion formula is the difference in equivalent heat flux between the two channels after compensation. When the heat flux difference is too small, the denominator approaches zero, and the calculation formula exhibits an ill-conditioned solution. Even minute temperature measurement noise and acquisition errors are amplified, causing drastic jumps and numerical distortion in the core temperature calculation results, rendering them completely worthless. Therefore, it is necessary to add a heat flux difference threshold judgment logic to promptly lock the data credibility when the effective heat flux difference is insufficient, using historical reliable values ​​to replace distorted calculation values, avoiding misleading judgments from abnormal temperature outputs, and improving the numerical stability and fault tolerance protection mechanism of the inversion algorithm. For example, the absolute difference between the equivalent heat flux of the fast channel and the equivalent heat flux of the slow channel after dynamic correction and environmental boundary compensation is calculated and compared with a preset difference threshold. When the difference is less than the preset difference threshold, it is determined that the current effective heat flux difference between the two thermal channels is insufficient, the denominator of the core temperature inversion formula is too small leading to an ill-conditioned solution, the inversion results are highly sensitive to small temperature measurement errors, and the reliability of the calculated values ​​decreases significantly. At this point, the system marks the transient core temperature of the human body obtained in this calculation as a low-confidence output marker. Instead of directly using the distorted value calculated in this calculation, it retrieves and uses the transient core temperature of the human body that has converged iteratively and has a reliable state at the most recent sampling time as the final effective output temperature at the current sampling time, ensuring that the entire set of core temperature measurement results are continuous, stable and usable.

[0042] In this embodiment, fast and slow heat conduction modules are attached to human skin to construct fast and slow channels, respectively. A thermal insulation layer is placed on the side of the fast and slow heat conduction modules away from the skin. The effective thermal conductivity of the fast and slow channels is calculated based on the thickness and cross-sectional area of ​​the fast and slow heat conduction modules. Based on the thermal insulation layer, the baseline temperature information of the fast and slow channels is collected. A natural transient identification stage is determined based on the synchronous change characteristics of the baseline temperature information. In the natural transient identification stage, dual-channel dynamic consistency residuals and environmental residuals are constructed based on the baseline temperature information, and the time constants of the fast and slow channels and the environmental heat transfer coefficient are estimated. Based on the time constants of the fast and slow channels, dual-timescale transient corrections are performed on the fast and slow channels to obtain the equivalent corrected temperature difference for each channel. The equivalent heat flux for each channel is reconstructed by combining the effective thermal conductivity of the fast and slow channels. Using the environmental heat transfer coefficient and baseline temperature information, the environmental reference heat flux corresponding to the fast and slow channels is calculated respectively. The difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel is used to obtain the environmental boundary disturbance for each channel. The environmental boundary disturbance is compensated by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel, and a transient core temperature inversion formula is established to calculate the human transient core temperature. The difference between the compensated fast channel equivalent heat flux and the slow channel equivalent heat flux is calculated. When the difference is less than a preset threshold, the human transient core temperature is marked as a low confidence state, and the most recently sampled reliable human transient core temperature is selected as the valid human transient core temperature at the current moment. Using the above method, the human transient core temperature is solved based on the identification of thermal parameters of fast and slow dual-channel heat conduction, dual-timescale thermal hysteresis correction, and environmental heat dissipation disturbance compensation. The confidence level of the ill-conditioned calculation results is marked by the dual-channel heat flux difference threshold, and historical confidence value is used as a fallback for fault-tolerant output to ensure the stability of the temperature measurement values ​​and the reliability of the results.

[0043] Example 3 Figure 5 This is a schematic diagram of the structure of the rapid human core temperature estimation device based on dual timescale heat flow provided in Embodiment 3 of the present invention, as shown below. Figure 5 As shown, the device includes: The construction module 310 is used to attach the fast heat conduction module and the slow heat conduction module to human skin, construct the fast channel and the slow channel respectively, and set the heat insulation control layer on the side of the fast heat conduction module and the slow heat conduction module away from the skin. The effective thermal conductivity corresponding to the fast channel and the slow channel is calculated according to the thickness and cross-sectional area of ​​the fast heat conduction module and the slow heat conduction module respectively. The estimation module 320 is used to collect the basic temperature information of the fast channel and the slow channel based on the thermal insulation control layer, determine the natural transient identification stage according to the synchronous change characteristics of the basic temperature information, and construct the dual-channel dynamic consistency residual and environmental residual according to the basic temperature information in the natural transient identification stage, and estimate the fast channel time constant, the slow channel time constant and the environmental heat transfer coefficient. The correction module 330 is used to perform dual-time-scale transient correction on the fast channel and the slow channel according to the fast channel time constant and the slow channel time constant respectively to obtain the equivalent corrected temperature difference corresponding to each channel, and reconstruct the equivalent heat flux corresponding to each channel by combining the effective thermal conductance corresponding to the fast channel and the slow channel. The compensation calculation module 340 is used to calculate the environmental reference heat flux corresponding to the fast channel and the slow channel respectively using the environmental heat transfer coefficient and the basic temperature information. The difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel is used to obtain the environmental boundary disturbance amount corresponding to each channel. The environmental boundary disturbance amount is compensated by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel and to establish the transient core temperature inversion relationship to calculate the transient core temperature of the human body.

[0044] The method and apparatus for rapid estimation of human core temperature based on dual-timescale heat flux provided in this embodiment constructs fast and slow channels by attaching fast and slow heat conduction modules to human skin. A heat insulation control layer is placed on the side of the fast and slow heat conduction modules away from the skin. The effective heat conduction corresponding to the fast and slow channels is calculated based on the thickness and cross-sectional area of ​​the fast and slow heat conduction modules, respectively. Based on the heat insulation control layer, the basic temperature information of the fast and slow channels is collected. A natural transient identification stage is determined based on the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, dual-channel dynamic consistency residuals and environmental residuals are constructed based on the basic temperature information, and the time constants of the fast and slow channels are estimated. The system employs a dual-timescale transient correction method based on the environmental heat transfer coefficient and the fast and slow channel time constants to obtain the equivalent corrected temperature difference for each channel. This is combined with the effective thermal conductance of the fast and slow channels to reconstruct the equivalent heat flux for each channel. Using the environmental heat transfer coefficient and baseline temperature information, the system calculates the environmental reference heat flux for both the fast and slow channels. The difference between the equivalent heat flux and the environmental reference heat flux for each channel is used to obtain the environmental boundary disturbance for each channel. A preset compensation coefficient is used to compensate for this disturbance, resulting in the compensated equivalent heat flux for each channel. A transient core temperature inversion formula is then established to calculate the transient core temperature of the human body. By combining a fast and slow dual-channel heat conduction structure with an external thermal insulation control layer, identifying dynamic heat transfer parameters using dual-channel residuals during the skin-contact transient phase, and reconstructing the equivalent heat flux using dual-timescale correction, along with compensation for environmental boundary disturbances, the system achieves direct inversion of the transient core temperature under dynamic human body conditions. This allows for high-precision body temperature estimation without waiting for a steady-state heat transfer state.

[0045] Based on the above embodiments, the building module includes: The module material acquisition unit is used to acquire the thermal conductivity, thickness and cross-sectional area of ​​the fast thermal conductivity module and the slow thermal conductivity module respectively, based on the materials used to manufacture them. An equivalent thermal resistance calculation unit is used to calculate the equivalent thermal resistance of the fast thermal conduction module and the slow thermal conduction module respectively based on their thermal conductivity, thickness and cross-sectional area. The effective thermal resistance calculation unit is used to invert the equivalent thermal resistances of the fast thermal conduction module and the slow thermal conduction module to obtain the effective thermal conductances of the fast thermal conduction module and the slow thermal conduction module, respectively.

[0046] Based on the above embodiments, the estimation module includes: The basic temperature information acquisition unit is used to collect basic temperature information of the fast channel and the slow channel based on the heat insulation control layer. The basic temperature information includes the skin-contact temperature of the fast channel, the skin-contact temperature of the slow channel, the ambient temperature of the fast channel, and the ambient temperature. The rate of change calculation unit is used to calculate the rate of temperature change of the fast channel skin-contact temperature and the slow channel skin-contact temperature relative to the previous sampling time, respectively. The natural transient identification stage determination unit is used to determine that the fast heat conduction module and the slow heat conduction module have completed human body contact when the temperature change directions of the fast channel skin-contact side and the slow channel skin-contact side are consistent, and the temperature change rate of the fast channel skin-contact side and the temperature change rate of the slow channel skin-contact side exceed the preset change rate threshold and continue to maintain the preset sampling period. The unit determines the current moment as the natural contact moment and uses the natural contact moment as the natural transient identification stage. An initial candidate example construction unit is used to determine the physical feasible region based on the materials used to manufacture the fast heat conduction module and the slow heat conduction module, the thickness of the heat insulation control layer, and the thermal conductivity of the heat insulation control layer. Based on the physical feasible region, initial candidate particles are constructed. The initial candidate particles include fast channel candidate time constant, slow channel candidate time constant, dual channel candidate gain ratio coefficient, and candidate environment heat transfer coefficient. The axial temperature difference change rate calculation unit is used to calculate the axial temperature difference of the fast channel, the axial temperature difference of the slow channel, the axial temperature difference change rate of the fast channel, and the axial temperature difference change rate of the slow channel based on the basic temperature information. A dual-channel dynamic consistency residual construction unit is used to construct a dual-channel dynamic consistency residual by utilizing the axial temperature difference of the fast channel, the axial temperature difference of the slow channel, the rate of change of the axial temperature difference of the fast channel, and the rate of change of the axial temperature difference of the slow channel. The candidate equivalent heat flux construction unit is used to construct the candidate equivalent heat flux corresponding to the fast channel and the slow channel respectively based on the effective thermal conductance, the axial temperature difference change rate of the fast channel and the axial temperature difference change rate of the slow channel. The ambient temperature calculation unit is used to calculate the ambient temperature using the basic temperature information and the cross-sectional areas of the fast and slow heat conduction modules. An environmental residual building unit is used to weightedly fuse candidate equivalent heat fluxes corresponding to the fast and slow channels, and to construct environmental residuals by combining ambient temperature and representative ambient temperature. The integrated residual construction unit is used to construct an integrated residual based on the dual-channel dynamic consistency residual and the environmental residual; The iterative estimation unit is used to iteratively update and correct the weights of the initial candidate particles using the comprehensive residual and the robust sequential Monte Carlo algorithm. After the iteration converges, the fast channel time constant, slow channel time constant, environmental heat transfer coefficient and dual-channel gain ratio coefficient are output.

[0047] Based on the above embodiments, the correction module includes: The equivalent corrected temperature difference calculation unit is used to perform dual-time-scale transient correction on the fast channel and the slow channel according to the fast channel time constant and the slow channel time constant to obtain the equivalent corrected temperature difference for each channel. The equivalent heat flux calculation unit is used to calculate the equivalent heat flux of each channel based on the equivalent corrected temperature difference corresponding to each channel, combined with the effective thermal conductance of the fast and slow channels.

[0048] Based on the above embodiments, the compensation calculation module includes: The fast-track ambient-side reference heat flux calculation unit is used to calculate the ambient-side reference heat flux of the fast track using the ambient heat transfer coefficient, ambient temperature, and fast-track ambient-side temperature. The slow-channel ambient-side reference heat flux calculation unit is used to calculate the ambient-side reference heat flux of the slow channel using the ambient heat transfer coefficient, ambient temperature, and slow-channel ambient-side temperature. The environmental boundary disturbance calculation unit is used to calculate the difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel to obtain the environmental boundary disturbance amount corresponding to each channel. The compensation processing unit is used to compensate for the environmental boundary disturbance by using a preset compensation coefficient to obtain the equivalent heat flux corresponding to each channel after compensation. The core temperature calculation unit is used to construct the transient core temperature inversion formula using the equivalent heat flux and basic temperature information corresponding to each channel after compensation, and to calculate the transient core temperature of the human body.

[0049] The dual-timescale heat flux rapid estimation device for human core temperature provided in this embodiment of the invention can execute the dual-timescale heat flux rapid estimation method for human core temperature provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0050] Example 4 Figure 6 This is a schematic diagram of the structure of a device provided in Embodiment 4 of the present invention. Figure 6 A block diagram is shown of an exemplary device 12 suitable for implementing embodiments of the present invention. Figure 6 The device 12 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0051] like Figure 6 As shown, device 12 is represented as a general-purpose computing device. Components of device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and a bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0052] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0053] Device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by device 12, including volatile and non-volatile media, removable and non-removable media.

[0054] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. Device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (… Figure 6 Not shown; usually referred to as a "hard drive"). Although Figure 6 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0055] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0056] Device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with device 12, and / or with any device that enables device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of device 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0057] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the rapid estimation method for human core temperature based on dual timescale heat flow provided in this embodiment of the invention.

[0058] Example 5 Embodiment 5 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a rapid estimation method for human core temperature based on dual-timescale heat flux as described in any of the above embodiments.

[0059] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0060] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0061] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including, but not limited to, wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0062] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0063] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A rapid method for estimating human core temperature based on dual-timescale heat flux, characterized in that, include: Fast and slow heat conduction modules are attached to human skin to construct fast and slow channels respectively. A heat insulation control layer is set on the side of the fast and slow heat conduction modules away from the skin. The effective thermal conductivity of the fast and slow channels is calculated based on the thickness and cross-sectional area of ​​the fast and slow heat conduction modules respectively. Based on the thermal insulation control layer, the basic temperature information of the fast channel and the slow channel is collected. The natural transient identification stage is determined according to the synchronous change characteristics of the basic temperature information. In the natural transient identification stage, the dual-channel dynamic consistency residual and environmental residual are constructed according to the basic temperature information, and the fast channel time constant, slow channel time constant and environmental heat transfer coefficient are estimated. Based on the fast channel time constant and the slow channel time constant, the fast channel and the slow channel are respectively subjected to dual time scale transient correction to obtain the equivalent corrected temperature difference of each channel, and the equivalent heat flux of each channel is reconstructed by combining the effective thermal conductance of the fast channel and the slow channel. Using the environmental heat transfer coefficient and base temperature information, the environmental reference heat flux corresponding to the fast channel and the slow channel is calculated respectively. The difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel is used to obtain the environmental boundary disturbance amount corresponding to each channel. The environmental boundary disturbance amount is compensated by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel. The transient core temperature inversion relationship is established, and the transient core temperature of the human body is calculated.

2. The method according to claim 1, characterized in that, The calculation of the effective thermal conductivity of the fast and slow channels based on the thickness and cross-sectional area of ​​the fast and slow thermal conduction modules respectively includes: Based on the materials used to manufacture the fast thermal conductivity module and the slow thermal conductivity module, the thermal conductivity, thickness, and cross-sectional area of ​​the fast thermal conductivity module and the slow thermal conductivity module are obtained respectively. The equivalent thermal resistance of the fast thermal conduction module and the slow thermal conduction module are calculated based on their thermal conductivity, thickness, and cross-sectional area, respectively. The calculation method for the equivalent thermal resistance of the fast thermal conduction module and the slow thermal conduction module is as follows: ； in, For the first Thermal conductivity of each heat-conducting module For the first The cross-sectional area of ​​each heat-conducting module, For the first The thickness of each heat-conducting module, For the first The equivalent thermal resistance of each heat-conducting module; The effective thermal conductance of the fast and slow thermal conduction modules is obtained by inverting their equivalent thermal resistances.

3. The method according to claim 1, characterized in that, The process, based on a thermal insulation control layer, involves collecting baseline temperature information for both fast and slow channels. The natural transient identification stage is determined based on the synchronous change characteristics of the baseline temperature information, including: Based on the heat insulation control layer, the basic temperature information of the fast channel and the slow channel is collected. The basic temperature information includes the skin-contact temperature of the fast channel, the skin-contact temperature of the slow channel, the ambient temperature of the fast channel, the ambient temperature of the slow channel, and the ambient temperature. Calculate the rate of temperature change of the fast channel skin-contact side temperature and the slow channel skin-contact side temperature relative to the previous sampling time, respectively. When the temperature changes in the fast channel skin-contact side and the slow channel skin-contact side in the same direction, and the temperature change rates of the fast channel skin-contact side and the slow channel skin-contact side exceed a preset change rate threshold and are maintained for a preset sampling period, it is determined that the fast heat conduction module and the slow heat conduction module have completed human body contact. The current moment is determined as the natural contact moment, and the natural contact moment is determined as the natural transient identification stage.

4. The method according to claim 3, characterized in that, The step of constructing dual-channel dynamic consistency residuals and environmental residuals based on the basic temperature information and estimating the fast-channel time constant, slow-channel time constant, and environmental heat transfer coefficient includes: The physical feasible region is determined based on the materials used in the fast and slow heat conduction modules, the thickness of the heat insulation control layer, and the thermal conductivity of the heat insulation control layer. Initial candidate particles are constructed based on the physical feasible region. The initial candidate particles include fast channel candidate time constant, slow channel candidate time constant, dual channel candidate gain ratio coefficient, and candidate environmental heat transfer coefficient. Based on the aforementioned basic temperature information, calculate the axial temperature difference of the fast channel, the axial temperature difference of the slow channel, the rate of change of the axial temperature difference of the fast channel, and the rate of change of the axial temperature difference of the slow channel, respectively. Using the axial temperature difference of the fast channel, the axial temperature difference of the slow channel, the rate of change of the axial temperature difference of the fast channel, and the rate of change of the axial temperature difference of the slow channel, a dual-channel dynamic consistency residual is constructed. This dual-channel dynamic consistency residual is expressed as: ； in, Let be the candidate time constant for the fast channel at sampling time k. Let be the rate of change of axial temperature difference in the fast channel at sampling time k. Let be the axial temperature difference of the fast channel at sampling time k. Let be the candidate gain scaling factor for the dual channels at sampling time k. Let be the candidate time constant for the slow channel at sampling time k. Let be the rate of change of axial temperature difference in the slow channel at sampling time k. Let k be the axial temperature difference of the slow channel at sampling time k. The dynamic consistency residual of the two channels at sampling time k; Based on the effective thermal conductance of the fast channel and the slow channel, the axial temperature difference change rate of the fast channel, and the axial temperature difference change rate of the slow channel, candidate equivalent heat fluxes for the fast channel and the slow channel are constructed respectively. The candidate equivalent heat fluxes for the fast channel and the slow channel are constructed as follows: ； in, For the first Candidate equivalent heat flux at sampling time k for each channel For the first Effective thermal conductivity of each channel For the first The axial temperature difference at each of the k sampling times in each channel. For the first Candidate time constants for k sampling times of each channel For the first The rate of change of axial temperature difference at k sampling times for each channel; Using the baseline temperature information and the corresponding cross-sectional areas of the fast and slow heat conduction modules, the representative ambient temperature is calculated as follows: ； in, Let k represent the ambient temperature at sampling time k. The cross-sectional area of ​​the fast heat conduction module, This refers to the cross-sectional area of ​​the slow-conducting heat module. Let k be the ambient temperature of the fast channel at sampling time k. Let k be the ambient temperature of the slow channel at sampling time k; The candidate equivalent heat fluxes corresponding to the fast and slow channels are weighted and fused, and combined with the ambient temperature and the representative temperature on the ambient side to construct the environmental residual. The environmental residual is constructed as follows: ； in, Let k be the environmental residual at sampling time k. Let be the heat transfer coefficient of the candidate environment at sampling time k. The cross-sectional area of ​​the fast heat conduction module, This refers to the cross-sectional area of ​​the slow-conducting heat module. Let k represent the ambient temperature at sampling time k. Let k be the ambient temperature at sampling time k. The weighted values ​​of the candidate equivalent heat fluxes for the fast and slow channels at sampling time k; Based on the dual-channel dynamic consistency residual and the environmental residual, a comprehensive residual is constructed, which is as follows: ； in, The combined residual at sampling time k, Let K be the dynamic consistency residual of the two channels at sampling time k. Let k be the environmental residual at sampling time k. and These are the dynamic consistency residual weights and the environmental residual weights, respectively. and These are the normal fluctuation scales corresponding to dynamic consistency residuals and environmental residuals, respectively. Using the comprehensive residual, the robust sequential Monte Carlo algorithm is used to iteratively update and correct the weights of the initial candidate particles. After the iteration converges, the fast channel time constant, slow channel time constant, environmental heat transfer coefficient and dual-channel gain ratio coefficient are output.

5. The method according to claim 4, characterized in that, The process of performing dual-timescale transient corrections on the fast and slow channels based on the fast and slow channel time constants respectively to obtain the equivalent corrected temperature difference for each channel, and reconstructing the equivalent heat flux for each channel by combining the effective thermal conductances of the fast and slow channels, includes: The equivalent corrected temperature difference for each channel is obtained by performing dual-time-scale transient corrections on the fast and slow channels based on the fast channel time constant and the slow channel time constant, respectively. The dual-time-scale transient correction formula is as follows: ； in, For the first The equivalent corrected temperature difference at each of the k sampling times of each channel. For the first The axial temperature difference at each of the k sampling times in each channel. To estimate the obtained first The time constant of each channel at k sampling times For the first The rate of change of axial temperature difference at k sampling times for each channel; Based on the equivalent corrected temperature difference for each channel, and combined with the effective thermal conductance of the fast and slow channels, the equivalent heat flux for each channel is calculated as follows: ； in, For the first Equivalent heat flux at k sampling times for each channel For the first Effective thermal conductivity of each channel For the first The equivalent corrected temperature difference at each of the k sampling times of each channel.

6. The method according to claim 5, characterized in that, The process involves calculating the environmental reference heat flux for the fast and slow channels using the environmental heat transfer coefficient and baseline temperature information, subtracting the equivalent heat flux for each channel from the environmental reference heat flux for each channel to obtain the environmental boundary disturbance for each channel, compensating the environmental boundary disturbance using a preset compensation coefficient, obtaining the compensated equivalent heat flux for each channel, establishing a transient core temperature inversion formula, and calculating the transient core temperature of the human body, including: The reference heat flux on the environmental side of the fast aisle is calculated using the environmental heat transfer coefficient, ambient temperature, and the ambient side temperature of the fast aisle. The calculation method is as follows: ； in, The ambient reference heat flux of the hot channel at sampling time k. The environmental heat transfer coefficient is estimated at sampling time k. The cross-sectional area of ​​the fast heat conduction module, Let k be the ambient temperature of the fast channel at sampling time k. Let k be the ambient temperature at sampling time k; The ambient reference heat flux of the slow channel is calculated using the ambient heat transfer coefficient, ambient temperature, and ambient side temperature of the slow channel. The calculation method is as follows: ； in, The environmental reference heat flux for the slow channel at sampling time k. The environmental heat transfer coefficient is estimated at sampling time k. This refers to the cross-sectional area of ​​the slow-conducting heat module. Let k be the ambient temperature of the slow channel at sampling time k. Let k be the ambient temperature at sampling time k; The environmental boundary disturbance of each channel is obtained by subtracting the equivalent heat flux of each channel from the environmental reference heat flux of each channel. The environmental boundary disturbance is compensated by a preset compensation coefficient to obtain the equivalent heat flux for each channel after compensation. The calculation method for the compensation process is as follows: ； in, The first sample after compensation at time k The equivalent heat flux of each channel, For the first Effective thermal conductivity of each channel For sampling time k, the first... The equivalent corrected temperature difference for each channel For the first Preset environmental boundary compensation coefficients for each channel. For sampling time k, the first... Environmental boundary disturbance of each channel; Using the equivalent heat flux and baseline temperature information corresponding to each compensated channel, a transient core temperature inversion formula is constructed to calculate the transient core temperature of the human body. The calculation method is as follows: ； in, Let k be the transient core temperature of the human body at sampling time k. Let k be the equivalent heat flux of the fast channel after compensation at sampling time k. The equivalent heat flux of the slow channel after compensation at sampling time k is given. For the fast-channel skin-contact temperature at sampling time k, The temperature on the skin-contact side of the slow channel at sampling time k.

7. The method according to claim 1, characterized in that, After the step of compensating for environmental boundary disturbances using a preset compensation coefficient, obtaining the equivalent heat flux corresponding to each channel after compensation, establishing the transient core temperature inversion formula, and calculating the transient core temperature of the human body, the method further includes: The difference between the compensated fast-channel equivalent heat flux and the slow-channel equivalent heat flux is calculated. When the difference is less than a preset threshold, the transient core temperature of the human body is marked as a low confidence state, and the most recently sampled reliable transient core temperature of the human body is selected as the effective transient core temperature of the human body at the current moment.

8. A device for rapid estimation of human core temperature based on dual timescale heat flux, characterized in that, include: The module is used to attach the fast heat conduction module and the slow heat conduction module to human skin, constructing fast channels and slow channels respectively, and setting a heat insulation control layer on the side of the fast heat conduction module and the slow heat conduction module away from the skin. The effective thermal conductivity corresponding to the fast channel and the slow channel is calculated based on the thickness and cross-sectional area of ​​the fast heat conduction module and the slow heat conduction module respectively. The estimation module is used to collect the basic temperature information of the fast channel and the slow channel based on the thermal insulation control layer, determine the natural transient identification stage according to the synchronous change characteristics of the basic temperature information, and construct the dual-channel dynamic consistency residual and environmental residual according to the basic temperature information and estimate the fast channel time constant, the slow channel time constant and the environmental heat transfer coefficient. The correction module is used to perform dual-time-scale transient correction on the fast channel and the slow channel according to the fast channel time constant and the slow channel time constant respectively to obtain the equivalent corrected temperature difference of each channel, and reconstruct the equivalent heat flux of each channel by combining the effective thermal conductance of the fast channel and the slow channel. The compensation calculation module is used to calculate the environmental reference heat flux corresponding to the fast channel and the slow channel respectively using the environmental heat transfer coefficient and the basic temperature information. The difference between the equivalent heat flux corresponding to each channel and the environmental reference heat flux corresponding to each channel is used to obtain the environmental boundary disturbance amount corresponding to each channel. The environmental boundary disturbance amount is compensated by a preset compensation coefficient to obtain the compensated equivalent heat flux corresponding to each channel and to establish the transient core temperature inversion relationship to calculate the transient core temperature of the human body.

9. A device, characterized in that, The device includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the rapid estimation method for human core temperature based on dual timescale heat flux as described in any one of claims 1-7.

10. A medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the rapid estimation method for human core temperature based on dual-timescale heat flux as described in any one of claims 1-7.

Images